Map of life expectancy at birth from Global Education Project.

Wednesday, March 11, 2020

Powers of Two

I have been skeptical of the cost-benefit balance of some of the more draconian measures to curtail the coronavirus epidemic. I still am -- indiscriminate travel bans and the isolation of communities or regions are very costly and aren't very effective. However, I am definitely not skeptical of the basic public health measures that are cost effective, including early diagnosis (which requires availability of adequate testing kits and elimination of obstacles to testing such as out of pocket cost), contact tracing, self-quarantine of mild cases and hospitalization with proper protection for health care workers of more severe cases, reduction or elimination of mass gatherings where there are outbreaks, and encouraging people to stay home if they feel sick, wash hands frequently, and avoid unnecessary close contact.

Taking these measures seriously, and fully implementing them, is actually a matter of urgency even though so far there have been only a bit more than 1,000 confirmed cases in the U.S. Liz Specht in Stat explains why. Right now, with an immunologically naive population, it appears the doubling time for cases is about six days. (Indeed, when Specht wrote this on March 5 there were 500 confirmed cases, so that's looking bang on.) Actually there are probably at least twice as many out there, so let's say our starting point as of March 11 is 2,000. (It's probably higher but let's be conservative.) So cases double 5 times in a month. That means we'll have, as a plausible expectation, 64,000 cases as of April 11, and more than a million by early May. Oh yeah, 2 million by May 10 or so and 4 million by mid-May. That's how powers of two work.

Sure, most people just have a cold and have to stay home for a week or so. But let's say 10% need to be hospitalized. Can the U.S. handle 200,000 additional people needing hospitalization? How about 400,000?

No.

If these assumptions are too pessimistic, the only difference it makes is that it takes us a week or two longer to get to catastrophe. If the hospitals are full, it means we can't take care of other needs either. Now maybe we get lucky and it turns out that warmer weather slows transmission, as with influenza. We don't know whether this is true, but even if it, we still need to worry about next fall. Since there's no vaccine, we can't stop more and more people from getting infected. But we can slow it down. If we can push more cases further into the future, then we have a better chance of being able to handle this. So while I do hope that public officials will think hard about being precise and proportionate in their actions, and weigh economic costs against likely benefits, I certainly do not argue for complacency. While your personal risk of severe illness is probably low (unless you're someone for whom it isn't), that isn't really the point.

No comments: